Of course not, radiation can only remove heat from the source. That’s why it accounts for whatever heat loss (missing energy) convection and radiation at the housing doesn’t. Maybe “heat loss” wasn’t a good choice of words.
Yeah, looking good!
How deep is that copper? Does the star overlap the copper and aluminum? It’s difficult to judge the size in the pictures.
I took a closeup of my C12 and noticed an acumulation dust and lint. I have plans to cut fins sometime soon but I think I’ll go with a wider groove than what I really want. You’re too fast.
The copper runs the length of the pill and is 5/8" diameter, coincidentally almost an exact fit for a 16mm star :) I cut the hole for a sliding fit, then press the crap out of it. It flows into the hole pretty good. I've successfully soldered two stars to similar, if smaller heat sinks prior to this and will be attaching this one the same way. It's surprisingly easy less than impossible. I was going to drive single XM-L2 at 7 amps, but am considering a MT-G2 instead at perhaps 9 amps. I don't know how the switch in this light will do in that regard.
One last pic for tonight. 007 will be happy to see the 4 jaw holding the head. The head has been reversed and faced to length. I bored this end to the diameter needed for the reflector and cut the bezel shoulder which has layout dye on it for threading. I don't thread when I'm tired, so that will be tomorrow.
Tofty is making a 10A+ switch if you’d like to ensure it’s equipped to handle the load. Nice solid switch, reverse clicky, should work very well for this light.
262g is just the pill. No decimal point. The head is still in the lathe chuck and I really don't want to have to put it back in and indicate it back to where it is now. I'm pretty slow at centering pieces in the 4 jaw. Anyway, I hesitate to estimate the mass of the head. It's just too hard to guess with all of the fins and two different diameters bored in it.
I don’t think I’d break the edges over, it looks perfect with the hard edges.
If you must, could you put an ever so slight bevel on them? A triangular file touched into the center of the groove while it’s spinning up to just barely knock the sharp edge off? Not sure what kind of carbide cutting edge might do the same, that would give you the ability to do each one exactly the same.
Really nice work you’re doing, I’m loving the piece I got today, precisely machined is always a sight to behold, and beholding it I will do!
Mrs and DBC, thanks for the compliments. Dale, I think that I will take your suggestion for breaking the edges ever so slightly with a file. I'll have to chuck it back up and hit it with the file when I get the bezel done.
I'll try to get more pictures as I go. Maybe some sort of assembly picture fest. I also took some video as I worked along, but somewhere along the line the camera lens got filthy. It's pretty bad, so I won't be posting it up.
The head and bezel are done. I may end up putting the bezel in the mill for some fancy work later, but I like it as it is for now.
Getting ready to cut threads
Head with bezel on, mounted on light. It's sitting on my project notebook in case you're wondering what all of the scribbling is.
Same shot without flash.
From the front, with pill and reflector installed.
Now I have to figure out what LED to put in it. I have some XM-L2's on copper stars lying around and one MT-G2 on an aluminum star (don't ask, thought it was on copper when I bought it). I have decided that for now I am going to screw whatever LED I go with to the pill so that it's easier to swap later if I want to.
I know that it is mainly a matter of esthetics, but a couple of comments on cooling: I agree with most of what is said above.
Dark colored hard anodized surfaces do radiate infrared well (We looked that up to check it when I was working on a junction box for a fighting vehicle.)
Fins help both convection and radiation cooling. They help convection by allowing air to go between the fins where it contacts more surface. They help radiation, if the surface is not already perfectly black to thermal IR, because the fins can reflect light radiated elsewhere in addition to radiating themselves. Trapezoidal (like an SK68) or triangular (probably uncomfortable to hold) fins cool better than square ones because the outside is more open to convection and radiation while the inside is thicker where heat conduction is needed to transfer heat out the fin. Square fins are also harder to machine, so we can assume that they are used for esthetics rather than for cooling. The fins in flashlights I have seen all look thicker than optimal. Probably one could improve the cooling by leaving more space between thinner fins.
There is also strength to consider. It resists crushing better if the fins are thick enough to resist buckling when the light is crushed. In general, fins increase hoop strength and reduce longitudinal strength, compared to the same weight as a cylinder, but if the wall is thin they may increase longitudinal strength also. Aircraft fuselages are finned on the inside to keep the monocoque skin from buckling.
The best examples of finned aluminum I have seen are the WW II air cooled airplane engines at the Smithsonian Air and Space Museum, mostly on the cylinders. They had forced air, so the gaps between fins could be deep and narrow.
Maybe that is one reason hobbyists make square fins, but isn’t the reason so many things have triangular screw threads that they are easier to cut than trapezoidal or square threads in mass production? It seems that with numerical control the same thing would apply to fins.
If the profile of a tool can cut a desired shape following just one simple path, why use a common tool and follow multiple or complex paths. I’ve considered grinding a tool to cut radiused, tapered grooves for fins which might look like it could’ve only been done by CNC but it would use the the same simple in/out/across path as required for cutting square fins. This is similar to the approach Bucket used to create a repeatable paraboloid requiring only one simple straight path. Likewise, I don’t see why cutting any type of threads would require multiple cuts but in either case, if you have a CNC and depending on how many you want to make, it might save on time and tooling.
